Purpose: Bacterial deoxyribonucleic acid (DNA) has been previously detected by polymerase chain reactions (PCR) in a significant percentage of culturally-sterile pediatric middle-ear effusions. The current study was designed to determine whether this represents the existence of viable bacteria or the persistence of residual DNA in the middle-ear cleft.
Materials and methods: The middle-ear cavities of two sets of chinchillas were inoculated with either: 1) 100 colony-forming units (CFU) of live Haemophilus influenzae, 2.2 x 10(6) CFU of pasteurized Moraxella catarrhalis, and 1000 ng of DNA (>10(8) genomic equivalents) from Streptococcus pneumoniae; or 2) 100 CFU of live S pneumoniae, 2.2 x 10(6) CFU of pasteurized M catarrhalis and 1000 ng of purified DNA from H influenzae. Animals were treated with ampicillin for 5 days beginning on day 3. A single-point longitudinal study design was used for sampling to eliminate the possibility of contamination.
Results: No DNA was detectable from the heat-killed bacteria or the purified DNA after day 3. However, DNA from the live bacteria persisted through day 21, even though all specimens were culture-negative following the initiation of antimicrobial therapy.
Conclusion: These findings indicate that purified DNA and DNA from intact but nonviable bacteria do not persist in the middle-ear cleft in the presence of an effusion, even following high copy inoculation. In contrast, antibiotic-treated bacteria persist in some viable state for weeks as evidenced by the differential ability of the PCR-based assay systems to detect the live bacteria, but not detect the heat-killed organisms.